CN100419664C - Incremental backup operations in storage networks - Google Patents

Incremental backup operations in storage networks Download PDF

Info

Publication number
CN100419664C
CN100419664C CNB2005101188875A CN200510118887A CN100419664C CN 100419664 C CN100419664 C CN 100419664C CN B2005101188875 A CNB2005101188875 A CN B2005101188875A CN 200510118887 A CN200510118887 A CN 200510118887A CN 100419664 C CN100419664 C CN 100419664C
Authority
CN
China
Prior art keywords
snapshot
difference file
snapshot difference
data
storage
Prior art date
Application number
CNB2005101188875A
Other languages
Chinese (zh)
Other versions
CN1770088A (en
Inventor
R·丹尼尔斯
L·纳尔逊
A·达尔曼
Original Assignee
惠普开发有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US10/979,395 priority Critical patent/US20060106893A1/en
Priority to US10/979395 priority
Application filed by 惠普开发有限公司 filed Critical 惠普开发有限公司
Publication of CN1770088A publication Critical patent/CN1770088A/en
Application granted granted Critical
Publication of CN100419664C publication Critical patent/CN100419664C/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation
    • G06F11/1402Saving, restoring, recovering or retrying
    • G06F11/1415Saving, restoring, recovering or retrying at system level
    • G06F11/1435Saving, restoring, recovering or retrying at system level using file system or storage system metadata
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation
    • G06F11/1402Saving, restoring, recovering or retrying
    • G06F11/1446Point-in-time backing up or restoration of persistent data
    • G06F11/1448Management of the data involved in backup or backup restore
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation
    • G06F11/1402Saving, restoring, recovering or retrying
    • G06F11/1446Point-in-time backing up or restoration of persistent data
    • G06F11/1448Management of the data involved in backup or backup restore
    • G06F11/1451Management of the data involved in backup or backup restore by selection of backup contents
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2053Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant
    • G06F11/2056Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant by mirroring
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2201/00Indexing scheme relating to error detection, to error correction, and to monitoring
    • G06F2201/84Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

Exemplary storage network architectures, data architectures, and methods for performing backup operations in storage networks are described. One exemplary method may be implemented in a processor in a storage network. The method comprises generating a snapclone of a source volume at a first point in time; contemporaneously activating a first snapdifference file logically linked to the snapclone; recording I/O operations that change a data set in the source volume to the first snapdifference file; closing the first snapdifference file; generating a backup copy of the snapclone at a second point in time, after the first point in time; and generating a backup copy of the first snapdifference file at a third point in time, after the second point in time.

Description

Carry out the method for backup operation in the storage networking

Technical field

Described theme relates to electronics and calculates, and more particularly, relates to the incremental backup operations in the storage networking.

Background technology

The ability of duplicating with the content of store storage device is a key character of storage system.Data can be stored concurrently, in case single memory storage or medium are out of order.When first memory storage or medium were out of order, the copy of the data that comprise in second memory storage or the medium then can be retrieved by system.Duplicate the establishment that also helps the fixedly record of content when duplicating with the ability of the content of store storage device.The previous version of the data that this feature permission user recovery is edited unintentionally or wiped.

Exist and the space and the processing cost of duplicating with the relevance of store storage device.For example, some memory storages hold within it and can't accept I/O (I/O) operation when just being replicated.In addition, the storage space that is used for preserving copy can't be used for other storage demand.

Storage system and storing software product can provide the mode of making the time point copy of coiling.In the part of these products, copy does not use the application of coiling and can not disturb significantly as quick as thought.In other products,, can make copy save the space by sharing storage rather than duplicating and all coil data.

But the known method that is used for the copy data file comprises restriction.The known dish clone method of a part does not provide quick copy.Other known dish clone method solutions are not saved the space.Other known dish clone method provides fast and saves the snapshot in space, but is not to carry out this generic operation in scalable, distributed, tabledriven virtual storage system.Therefore, in memory storage, still need improved replicate run.

Summary of the invention

In a demonstration realized, a kind of computing method can realize in the processor in storage networking.This method comprises: the snapshot clone who produces source book at very first time point; Activate the first snapshot difference file that logically is linked to the snapshot clone simultaneously; To change the I/O operation note of the data set in the source book in the first snapshot difference file; Close the first snapshot difference file; After very first time point, produce snapshot clone's backup copies at second time point; And after second time point, produce the backup copies of the first snapshot difference file at the 3rd time point.

Description of drawings

Fig. 1 is an illustrative that demonstration realizes utilizing the networking computing system of storage networking.

Fig. 2 is an illustrative that demonstration realizes of storage networking.

Fig. 3 is an illustrative that demonstration realizes that can be used for realizing the calculation element of main frame.

Fig. 4 is an illustrative that demonstration realizes of storage unit.

Fig. 5 illustrates that the exemplary memory of LUN represents.

Fig. 6 is the illustrative of the data allocations in virtual storage system.

Fig. 7 is the illustrative that is used for realizing at storage networking an example data architecture of snapshot difference file.

Fig. 8 is the illustrative that is used for creating and using at storage networking a demonstration file structure of snapshot difference file.

Fig. 9 a-9b is the illustrative of the memory assignment map of snapshot difference file.

Figure 10 is a process flow diagram, and the operation of the demonstration methods that is used for creating the snapshot difference file is described.

Figure 11 is a process flow diagram, and the operation that is used for carrying out the demonstration methods of read operation under the environment that utilizes one or more snapshot difference file is described.

Figure 12 is a process flow diagram, and the operation that is used for carrying out the demonstration methods of write operation under the environment that utilizes one or more snapshot difference file is described.

Figure 13 is a process flow diagram, illustrates to be used for the snapshot difference file is merged to operation in the demonstration methods of Logical Disk.

Figure 14 is a process flow diagram, and the operation in the demonstration methods that utilizes the snapshot difference file in recovery operation is described.

Figure 15 is a process flow diagram, and an operation that demonstration realizes of the method that is used for managing automatically backup operation is described.

Embodiment

As herein described is exemplary storage network architecture, data volume architecture and the method that is used for creating and using at storage networking difference file.Method as herein described can be presented as the logical order in the computer-readable medium.When carrying out on processor, logical order makes the general-purpose computations device be programmed to realize the custom-built machine of described method.When being configured to carry out method as herein described by logical order, processor is configured for carrying out the structure of described method.

The exemplary network architecture

Theme as herein described can realize in the storage architecture that the virtual data storage on system-level is provided, and makes virtually to realize in SAN.In realization as herein described, utilize the computing system of memory storage to be called main frame.In a kind of typical case realized, main frame was to represent itself or representative to be coupled to any computing system of the system consumption data storage resource capacity of main frame.For example, main frame can be the supercomputer of handling big database, transaction server of care of transaction record or the like.Perhaps, main frame can be the file server in Local Area Network or the wide area network (WAN), and it provides stores service for enterprise.

In the Direct Attached Storage solution, this main frame can comprise one or more disk controllers or the RAID controller that is configured to manage a plurality of direct-connected disk drives.By contrast, in SAN, main frame as one man is connected to SAN via the fiber channel in high speed interconnection technique, for example instantiation (FC) switching fabric.

Virtual SAN architecture comprises one group of storage unit, and wherein, each storage unit comprises the memory storage pond that is called the dish group.Each storage unit comprises the parallel memory controller that is coupled to the dish group.Memory controller adopts the fiber channel arbitrated loop to connect or arrives memory storage by the network coupled such as the fiber channel switching fabric.Memory controller also can connect by point-to-point and intercouple, make they can coordinated management memory capacity providing the computing machine that uses memory capacity.

Network architecture as herein described is represented distributed computing environment, for example adopts the enterprise computing system of proprietary SAN.But network architecture can be easy to convergent-divergent up and down, so that satisfy the needs of application-specific.

Fig. 1 is an illustrative that demonstration realizes utilizing the networking computing system 100 of storage networking.In a demonstration realized, memory storage pond 110 can be embodied as virtual memory storage pond, and is for example described in people's such as Lubbers the U.S. Patent Application Publication No. of having announced 2003/0079102, by reference that this is open intactly incorporated herein.

A plurality of Logical Disk (being called logical block or LUN again) 112a, 112b can distribute in memory storage pond 110.Each LUN 112a, 112b comprise the logical address of a series of adjacency, and they can come addressing by asking to be mapped to uniquely identified LUN 112a, 112b from the employed connection protocol of host apparatus by host apparatus 120,122,124 and 128.Main frame, can provide service to other calculating or data handling system or device as server 128.For example, client computer 126 can be via main frame, as server 128 access to storage device ponds 110.Server 128 can provide file service to client computer 126, and other service such as transaction services, E-mail service can be provided.Therefore, customer set up 126 may or may directly not use the memory storage that is consumed by main frame 128.

For example wireless device 120 devices such as grade and the computing machine 122,124 that also can be used as main frame can logically be directly coupled to LUN 112a, 112b.Main frame 120-128 can be coupled to a plurality of LUN 112a, 112b, and LUN 112a, 112b can share between a plurality of main frames.Each comprised storer in the device shown in Figure 1, mass storage device and the data-handling capacity to a certain degree that is enough to the supervising the network connection.

For example LUN such as LUN 112a, 112b comprise one or more redundant memories (RStore), and they are base units of reliable memory.RStore comprises the ordered set of the physical store section (PSEG) with associated redundant attribute, and whole being included in the single redundant memory collection (RSS).With the heritage storage system analogy, PSEG is similar to disk drive, and each RSS is similar to the RAID saveset that comprises a plurality of drivers.

The PSEG that realizes specific LUN can distribute on any amount of physical store dish.In addition, the physical store capacity of specific LUN 102 expressions can be configured to realize providing the various storage classes of variation capacity, reliability and availability aspect.For example, some LUN can represent the storage of band, mirror image and/or parity checking protection.Other LUN can represent not dispose the memory capacity of band, redundance or parity checking protection.

In a demonstration realized, RSS comprised the subclass of the physical disks in the logical device allocation domain (LDAD), and can comprise six to 11 phisical drives (can dynamically change).Phisical drive can have different capacity.In order to shine upon, but the phisical drive allocation index among the RSS (for example 0,1,2 ..., 11), and, can be organized into (being adjacent odd and even number index) for RAID-1.Comprise that the problem that the big RAID volume of many dishes exists is, coiling out of order possibility obviously increases along with adding more multiple driver.For example, can to run into the possibility of driver malfunction (perhaps more simultaneously two driver malfunctions) be the twice of eight drive systems to 16 drive systems.According to the present invention, because data protection launches in certain RSS, and not to stride a plurality of RSS, therefore, the dish fault among RSS is for the not influence of availability of other any RSS.Therefore, the RSS that realizes data protection must stand twice driver malfunction in RSS rather than twice fault in the total system.Because the group during RAID-1 realizes is right, therefore not only two drivers must be out of order in specific RSS, and in the driver among the RSS specific one must be second and be out of order (promptly second out of order driver must be paired with first out of order driver).Saveset has improved performance, reliability and the availability of the data of total system to this atomization of a plurality of RSS that wherein can manage each RSS independently.

San management device equipment 109 is coupled to management logic Pan Ji (MLD) 111, and it is a metadata container of describing logical organization and other logical organization that system uses be used to create LUN 112a, 112b, LDAD 103a, 103b.The part of available physical store capacity keeps as legal space 113 in the memory storage pond 101, can not distribute to LDAD 103a, 103b, therefore can't be used to realize LUN 112a, 112b.In a particular instance, each physical disks that adds memory storage pond 110 has the capacity of the reservation quantity that can be designated as legal space 113 (" n " individual physical sector for example).MLD 111 is formed mirror image in this legal space of a plurality of phisical drives, even therefore also can be by access when driver is out of order.In a particular instance, at least one phisical drive related with each LDAD 103a, 103b comprises the copy (called after " method fixed driver ") of MLD 111.San management equipment 109 may wish related such as LDAD 103a, 103b and LUN 112a, 112b the title string and the information of the markers of object birthday etc.For the ease of this behavior, administration agent adopts MLD 111 to store this information as metadata.MLD 111 implicit expression when the establishment of each LDAD103a, 103b is created.

For example, legal space 113 is used for storing the information that comprises physical storage ID (unique ID of each phisical drive), Version Control information, type (legal/non-legal), RSS ID (identify this dish and belong to which RSS), RSS skew (identifying the relative position of this dish in RSS), storage unit ID (identify this dish and belong to which storage unit), PSEG size and shows whether this dish is the status information of method price fixing.This metadata PSEG also comprises the PSEG free list of whole physical storage, may take the form of assignment bit map.In addition, legal space 113 comprises the PSEG assignment record (PSAR) of each PSEG on the physical disks.PSAR comprises the indication of the RSD under PSAR signature, metadata version, PSAR operating position and this PSEG.

CSLD 114 is metadata containers of another kind of type, comprising distribute the address space in each LDAD103a and 103b but differently can cross over the logical drive of a plurality of LDAD 103a and 103b with LUN 112a and 112b.Each LDAD 103a, 103b preferably comprise the space of distributing to CSLD 114.CSLD 114 preserves the metadata of the logical organization of describing given LDAD103, comprising main logic dish metadata container (PLDMC), it comprises the array of the descriptor (being called RSDM) that is described in each the LUN 112a, employed each RStore of 112b that realize among LDAD 103a, the 103b.CSLD 114 realize being generally used for such as dish create, evenly adjust, RSS merges, RSS is cut apart and regenerate the metadata of task.This metadata comprises the status information of each physical disks, it shows that physical disks is that " normally " (promptly mode is worked according to expectation), " losing " (promptly unavailable), " merging " (promptly reappeared and before use must the standardized driver of losing), " replacement " (be that driver is marked as and removes, and data must copy on the distributed spare part) and " regeneration " is (promptly, driver is unavailable, and requires its data reproduction to distributed spare part).

Logical Disk catalogue (LDDIR) data structure among the CSLD 114 is the whole LUN 112a among any LDAD103a, the 103b, the catalogue of 112b.The RSD of position that clauses and subclauses among the LDDS comprise general unique ID (UUID) and represent the main logic dish metadata container (PLDMC) of that LUN 102.RSD points to the pointer of basic RSDM or the entrance of corresponding LUN 112a, 112b.Like this, thus the specific metadata of concrete LUN 112a, 112b can visit by indexing the basic RSDM that LDDIR searches specific LUN 112a, 112b.PLDMC (mapping structure for example described below) but in the metadata load memory realize concrete LUN 112a, 112b.

Therefore, the metadata of the various ways that can be used for recovering is realized in memory storage pond shown in Figure 1.CSLD 111 realize being generally used for such as dish create, evenly adjust, RSS merges, RSS is cut apart and regenerate the metadata of task.The PSAR metadata of preserving comprises the more metadata of citation form in the known location on each dish, and it is not mapped in the storer, but can be when needed visits from its known location, thus all metadata in the regenerative system.

Each comprised storer in the device shown in Figure 1, mass storage device and the data-handling capacity to a certain degree that is enough to the supervising the network connection.Computer program device according to the present invention realizes in the storer of various devices shown in Figure 1, and is enabled by the data-handling capacity of device shown in Figure 1.

In a demonstration realized, each LDAD 103a, 103b can be with few corresponding to four disk drives, the thousands of disk drives of as many as.In particular instance, require minimum eight drivers of each LDAD, so that support to adopt four RAID-1 among LDAD 103a, the 103b that coils in pairs.LUN 112a, the 112b that defines among LDAD 103a, the 103b can represent that number megabyte or following memory block are up to 2 terabytes or above memory block.Therefore, hundreds of or thousands of LUN 112a, 112b can define in given LDAD 103a, 103b, thereby serve a large amount of storage demands.Like this, large enterprises can provide service by the single memory storage pond 1101 of the shared memory in isolated memory area that each workstation that is exclusively used in the enterprise is provided and the whole enterprise.In addition, enterprise can realize that a plurality of LDAD 103a, 103b and/or a plurality of memory storages pond 1101 provide unlimited virtually memory capacity.Therefore, logically, provide big dirigibility in configuration and the visit according to the virtual storage system of this explanation.

Fig. 2 can be used to realize for example illustrative of the exemplary storage network 200 in the memory storage pond in memory storage pond 110.Storage networking 200 comprises a plurality of storage unit 210a, 210b, the 210c that connects by communication network 212.Storage unit 210a, 210b, 210c can be embodied as one or more memory storages that can connect communicatedly.The exemplary storage device comprises can be to Hewlett-Packard Corporation (Palo Alto, California, USA) the STORAGEWORKS line of the memory storage of Gou Maiing.Communication network 212 can be embodied as proprietary dedicated network, for example fiber channel (FC) switching fabric.Perhaps, the several portions of communication network 212 can adopt according to for example public communication network of the suitable communication protocol of internet small computer serial line interface (iSCSI) agreement and realize.

Client computer 214a, 214b, 214c can be by main frames, visit storage unit 210a, 210b, 210c as server 216,220.Client computer 214a, 214b, 214c can be directly or via network 218, be connected to file server 216 as Local Area Network or wide area network (WAN).The quantity that can be included in storage unit 210a, 210b in any storage networking, 210c mainly is subjected to the restriction of the connectedness that realizes in the communication network 212.For instance, the switching fabric that comprises single FC switch can interconnect 256 or multiport more, thereby the possibility of hundreds of storage unit 210a, 210b, 210c is provided in the single storage networking.

Main frame 216,220 is embodied as server computer usually.Fig. 3 is the illustrative that can be used for realizing the demonstration calculation element 330 of main frame.Calculation element 330 comprises one or more processors or processing unit 332, system storage 334 and will comprise that the various system components of system storage 334 are coupled to the bus 336 of processor 332.In the bus structure of the some types of bus 336 expression any one or a plurality of is comprising the memory bus or Memory Controller, peripheral bus, Accelerated Graphics Port and processor or the local bus that adopt in the various bus architectures any.System storage 334 comprises ROM (read-only memory) (ROM) 338 and random-access memory (ram) 340.Comprising the basic input/output (BIOS) 342 that transmits the basic routine of information in start-up course between the element that for example helps in the calculation element 330 is stored among the ROM 338.

Calculation element 330 also comprises and is used for hard disk drive 344 that the hard disk (not shown) is read and writes, and can comprise the disc driver 346 that moveable magnetic disc 348 is read and writes and be used for the CD drive 350 that reads or write to removable CD 352, as CD ROM or other light medium.Hard disk drive 344, disc driver 346 and CD drive 350 are connected to bus 336 by scsi interface 354 or certain other suitable interface.Driver and related computer-readable medium thereof provide non-volatile memories to computer-readable instruction, data structure, program module and other data for calculation element 330.Though exemplary environment as herein described adopts hard disk, moveable magnetic disc 348 and removable CD 352, the computer-readable medium of other type such as magnetic tape cassette, flash card, digital video disk, random-access memory (ram), ROM (read-only memory) (ROM) also can be used for exemplary operational environment.

Many program modules can be stored among hard disk 344, disk 348, CD 352, ROM 338 or the RAM 340, comprising operating system 358, one or more application program 360, other program module 362 and routine data 364.The user can will order and information input computing device 330 by for example input media such as keyboard 366 and indicating device 368.Other input media (not shown) can comprise microphone, operating rod, game control pad, satellite dish, scanner or the like.These and other input media is connected to processing unit 332 by the interface 370 that is coupled to bus 336.The display device of monitor 372 or other type also is connected to bus 336 via interfaces such as for example video adapters 374.

Calculation element 330 can be worked in the networked environment that uses one or more remote computers, connects as the logic of remote computer 376.Remote computer 376 can be personal computer, server, router, network PC, peer or other common network node, and generally include above many or whole in the calculation element 330 described elements, but memory storage apparatus 378 only has been described in Fig. 3.Logic shown in Figure 3 connects and comprises LAN380 and WAN 382.

When being used for the LAN networked environment, calculation element 330 is connected to LAN (Local Area Network) 380 by network interface or adapter 384.When being used for the WAN networked environment, calculation element 330 generally includes modulator-demodular unit 386 or is used at wide area network 382, as setting up other parts of communication on the Internet.Can be that internal or external modulator-demodular unit 386 is connected to bus 336 via serial port interface 356.In networked environment, can be stored in the remote memory storage device for calculation element 330 or the described program module of its several portions.It is exemplary that network shown in should be known in connects, and can adopt other method of the communication link of setting up between the computing machine.

Main frame 216,220 can comprise the host adapter hardware and software, thereby is implemented to the connection of communication network 212.Connection to communication network 212 can be depended on bandwidth requirement by optically-coupled or more traditional conduction cable.Host adapter can be embodied as the insertion card on the calculation element 330.Main frame 216,220 can be realized any amount of host adapter, thereby the connection to communication network 212 of the quantity that hardware and software supports is provided.

In general, the data processor of calculation element 330 is programmed by the instruction in the various computer-readable recording mediums that are stored in computing machine at different time.For example, program and operating system can distribute on floppy disk, CD-ROM or in electric mode, and the supplementary storage of the installation or the computing machine of packing into.During operation, program to small part is loaded into the main electronic memory of computing machine.

Fig. 4 is an illustrative that demonstration realizes of storage unit 400 that can be used to realize the storage unit of for example 210a, 210b or 210c.With reference to Fig. 4, storage unit 400 comprises and two NSCs (NSC) 410a, the 410b that are called disk array controller manages operation and transmission to the data of one or more disk drives 440,442.NSC 410a, 410b can be embodied as the insertion card with microprocessor 416a, 416b and storer 418a, 418b.Each NSC 410a, 410b comprise two host adapter port 412a, 414a, 412b, the 414b that promptly is provided to the interface of main frame by the communication network of for example switching fabric etc.In fiber channel was realized, host adapter port 412a, 412b, 414a, 414b can be embodied as FC N port.Each host adapter port 412a, 412b, 414a, 414b administrative login and with the interface of switching fabric, and in login process, be assigned with structure unique port ID.Architecture shown in Figure 4 provides the fully redundance storage unit; Only require that single NSC realizes storage unit.

Each NSC 410a, 410b also comprise 438 communication port 428a, the 428b of communicating to connect that realizes between NSC 410a, the 410b.Communicate to connect 438 and can be embodied as the connection of FC point-to-point, perhaps according to other any suitable communication protocol.

In a demonstration realized, NSC 410a, 410b also comprised and realize a plurality of FCAL port 420a-426a, the 420b-426b that communicate to connect with a plurality of memory storages, as the fiber channel arbitrated loop (FCAL) of array of disk drives 440,442.Though described embodiment realizes being connected with the FCAL of array of disk drives 440,442, everybody will appreciate that, can adopt other communication protocol to realize with communicating to connect of array of disk drives 440,442.For example, can adopt connection of FC switching fabric or small-size computer serial line interface (SCSI) rather than FCAL configuration.

In operation, array of disk drives 440,442 memory capacity that provided can be added memory storage pond 110 to.When application need memory capacity, the logical order on the principal computer 128 from one or more storage site can with array of disk drives 440,442 on available memory capacity set up LUN.Everybody will appreciate that, because LUN is logical block and physical location not necessarily, the amount of physical memory that therefore constitutes LUN can be distributed on a plurality of storage unit.Among one or more LUN of data storage in storage networking that use.The application query principal computer that needs access data, principal computer retrieve data and data forwarding give used from LUN.

One or more storages that realize based on RAID among storage unit 210a in the storage networking 200,210b, the 210c.RAID (redundant array of independent disks) storage system is the disk array system that the part of wherein physical store capacity is used for the storage redundancy data.The RAID system is common be characterized as six kinds of architectures under acronym RAID, enumerating one of them.RAID 0 architecture is the disk array system that is configured to without any redundance.Because in fact this architecture is not redundant architecture, therefore, RAID 0 often omits from the argumentation of RAID system.

RAID 1 architecture relates to the memory disc according to the configuration of mirror image redundance.Original data storage is in one group of dish, and duplicate copies of data then is kept in the other dish.RAID 2 to RAID 5 architectures all relate to the parity check type redundant storage.Special concern be RAID 5 systems distributed data and parity information on a plurality of dishes.These dishes are divided into the address area of equal sizes usually, are called " piece ".Be called " bar " from each chunk that coils, has the same unit address realm.In RAID 5, the parity block that each bar has the data of N piece and comprises the redundant information of N the data in the piece.

In RAID 5, parity block is circulation one by one on the difference dish.For example, in RAID 5 systems with five dishes, the parity block of article one may be on the 5th dish; The parity block of second may be on the 4th dish; Article three, parity block may be on the 3rd dish; Or the like.The parity block of bar subsequently centers on disk drive " precession " with helicon mode (but other pattern also is feasible) usually.The difference of RAID 2 to RAID 4 architectures and RAID5 is their calculating and the mode of placing parity block on dish.The specific RAID class that is realized is unimportant.

Fig. 5 illustrates that the exemplary memory of LUN 112a, the 112b of a demonstration in realizing represents.Storer represents it mainly is a kind of mapping structure of realizing in the storer of NSC 410a, 410b, its realize the request expressed with LBA (Logical Block Addressing) (LBA) from main frame, as shown in Figure 1 main frame 128 at the physical disks driver, as the conversion of the read/write command of the specific part of disk drive 440,442.Wish that storer represents enough for a short time, so that be fit to suitable memory size, it is accessed at work to make that it can be easy to, and represents page-in and access the requirement of storer of NSC minimum or do not have for storer.

Storer as herein described represents to make each LUN 112a, 112b can realize from the memory capacity of 1 megabyte to 2 terabyte.Can imagine the bigger memory capacity of each LUN 112a, 112b.For convenience of explanation, adopt the maximal value of 2 terabytes in this explanation.In addition, storer represents to make each LUN 112a, 112b can adopt the RAID data protection of any kind to define, comprising multistage RAID protection and support without any redundance.In addition, polytype RAID data protection can realize in single LUN 112a, 112b, and the logical disc address (LDA) that makes the scope of winning is corresponding to protected data not, and second group of LDA among identical LUN112a, the 112b realizes RAID 5 protections.Therefore, realize this species diversity of the necessary processing flexibly of data structure that storer is represented, but still be effectively, make LUN 112a, 112b not need extra data structure.

The persistent copy that storer shown in Figure 5 is represented is kept among the PLDMDC of foregoing each LUN112a, 112b.When system reads the metadata that comprises in the legal space 113 so that obtain the pointer that points to corresponding PLDMDC and then retrieval PLDMDC and when loading the 2nd grade of mapping graph (L2MAP) 501, realize that the storer of specific LUN 112a, 112b is represented.This carries out for each LUN 112a, 112b, but in general operation, this will occur once when creating LUN 112a, 112b, and after this, storer is represented and will be retained in the storer when being used.

The LDA of appointment was mapped to the skew among specific RStore and the RStore during the Logical Disk mapping layer will be asked.With reference to the embodiment shown in Fig. 5, LUN can adopt L2MAP501, LMAP 503 and redundance set descriptor (RSD) 505 to realize as the main structure that logical disc address is mapped to the represented physical memory cell in address.Mapping structure shown in Figure 5 is realized for each LUN 112a, 112b.Single L2MAP handles whole LUN112a, 112b.Each LUN 112a, 112b are by a plurality of LMAP 503 expressions, and wherein the concrete quantity of LMAP503 depends on the actual address space of distributing any preset time.RSD505 also only exists the storage space that is distributed.Utilize this segmentation directory method, big storage volume, the structure as shown in Figure 5 of sparsely having loaded institute's allocate storage represented the memory block distributed effectively, makes the data structure minimum of unallocated memory block simultaneously.

L2MAP 501 comprises a plurality of clauses and subclauses, and each clauses and subclauses are wherein represented 2 GB of address space.Therefore, for 2 billion LUN 112a, 112b, L2MAP 501 comprises 1024 clauses and subclauses, so that cover the whole address space in the particular instance.Each clauses and subclauses can comprise the status information corresponding with the memory block of corresponding 2 GB and the pointer of corresponding LMAP descriptor 503.Status information and pointer are only just effective when the address space of corresponding 2 GB has been assigned with, and therefore, a part of clauses and subclauses among the L2MAP 501 will be for sky or invalid in many application.

The represented address realm of each clauses and subclauses among the LMAP 503 is called logical disc address allocation units (LDAAU).In specific implementation, LDAAU is 1 megabyte.Distribute LDAAU in LMAP 503, to create clauses and subclauses for each, and irrelevant with the actual utilization of memory block among the LDAAU.In other words, the big I of LUN 102 enlarges according to the increment of 1 megabyte or dwindles.LDAAU represents granularity, and the address space among LUN 112a, the 112b can adopt this granularity to distribute to the specific memory task.

LMAP 503 only to distribution address space each 2 GB increment exist.If the memory block less than 2 GB is used for specific LUN 112a, 112b, then only need a LMAP 503, and if the memory block of 2 terabytes is used, then will have 1024 LMAP 503.Each LMAP 503 comprises a plurality of clauses and subclauses, and each clauses and subclauses wherein are alternatively corresponding to redundant segments (RSEG).RSEG is the atom logical block, it is roughly similar to the PSEG in the physical domain-be similar to the Logical Disk subregion of RStore.In a particular embodiment, RSEG is the logical block of the memory block of a plurality of PSEG of leap and the data protection that realizes selected type.In preferred a realization, the whole RSEG among the RStore are tied to the LDA of adjacency.For the basic physics dish performance that keeps order to transmit, wish to locate all RSEG from the RStore purlieu in order, so that keep physical abutment according to the LDA space.But if physical resource becomes deficiency, the separated region that then may need to stride across LUN 102 is from the RStore RSEG that distributes.The logical disc address of appointment is selected the particular items among the LMAP 503 corresponding with specific RSEG in the request 501, and specific RSEG is corresponding to the 1 megabyte address space of distributing to specific RSEG#.Each LMAP clauses and subclauses also comprises status information relevant with specific RSEG and RSD pointer.

RSEG# can be omitted alternatively, and it makes RStore originally as assignable minimum atom logical block.The omission of RSEG# has reduced the size of LMAP clauses and subclauses, and the storer of permission LUN102 is represented the memory block needs memory resource still less for every megabyte.Perhaps, the big I of RSEG increases rather than omits fully the notion of RSEG, and it also relies on the granularity of the atom logical block of the memory block that reduces to reduce demand to memory resource.Therefore, change to satisfy the needs of application-specific with the big I process of the proportional RSEG of RStore.

The specific RSD 505 of RSD pointed, it comprises the metadata of the RStore that describes corresponding RSEG place.As shown in Figure 5, RSD comprises redundant memory collection selector switch (RSSS), wherein comprises redundant memory collection (RSS) sign, physics member selection and RAID information.It is the tabulation of the employed phisical drive of RStore in essence that the physics member selects.More in general RAID information be data protection information perhaps, describes any data protection type that realizes among the specific RStore.Each RSD also comprises a plurality of fields of the specific PSEG numbering in the driver that is identified at the physics member selection that physically realizes the respective stored capacity.Each PSEG# that lists is corresponding to listing one of member in the physics member selective listing of RSSS.Can comprise any amount of PSEG, still, In a particular embodiment, each RSEG adopts four to eight PSEG of the RAID type defined of being realized by RStore to realize.

In operation, specify LUN 112a, 112b and address for each request of memory block access.For example the NSC of NSC 410a, 410b etc. is mapped to specific LUN 112a, 112b with specified logical drive, when the L2MAP 501 of that LUN 102 also is present in storer it is not loaded in the storer then.All LMAP and the RSD of LUN 102 preferably also are loaded in the storer.Ask specified LDA to be used for indexing L2MAP501, it points to one specific among the LMAP again.Specified address is used for being determined to the skew of specifying LMAP in the request, so that return the specific RSEG corresponding with asking assigned address.In case RSEG# is known, then corresponding RSD is examined so that identify the member's who belongs to redundant segments specific PSEG and make NSC 410a, 410b can produce the driver particular command so that the metadata of access institute request msg.Like this, be easy to be mapped to must be by access so that realize one group of PSEG of given storage request for LDA.

L2MAP each LUN 112a, 112b consume 4 kilobyte, and irrelevant with the size in the demonstration realization.In other words, L2MAP comprises the clauses and subclauses that cover whole 2 terabyte maximum address scopes, even have only sub-fraction in fact to be assigned to LUN112a, 112b in that scope.Consideration can be adopted the L2MAP of variable-size, has few storer but this realization will increase complexity and saves.The every megabyte address space of LMAP section consumes 4 bytes, and the every megabyte of RSD then consumes 3 bytes.Different with L2MAP, LMAP section and RSD only exist for institute addresses distributed space.

Fig. 6 is the illustrative of the data allocations in virtual storage system.With reference to Fig. 6, redundant layer is selected PSEG 601 according to the expection protection and according to NSC data organization rule, and they are assembled to create redundant memory (RStore).The set of the PSEG that answers with specific redundancy storer set pair is called " RStore ".The data protection rule may require the PSEG among the RStore to be arranged in disk drive separately or the encapsulation that separates, and perhaps is in different geographic position.For example, basic RAID-5 rule supposition bar is cut data and is related to the bar that strides across some Standalone Drives and cut.But because each driver comprises a plurality of PSEG, therefore, redundant layer of the present invention guarantees that PSEG chooses from the driver that satisfies anticipatory data protecting standard and availability of data and performance standard.

RStore is intactly distributed to specific LUN 102.But the RStore subregion is 1 megabyte section (RSEG), as shown in Figure 6.80% of the physical disks capacity that each RSEG among Fig. 6 owing to only provide according to the considerable parity data of RAID 5 rale store is consumed.When being configured to RAID 5 savesets, each RStore will comprise four on the PSEG data and the parity information on the 5th PSEG (not shown) similar to the RAID4 memory block.The 5th PSEG does not constitute calm measuring angle and it seems the whole memory capacity of RStore with four PSEG.Stride across a plurality of RStore, parity checking will drop on the various drivers, so that RAID5 is provided protection.

RStore is the virtual address space (being 8 megabyte in the example) of fixed qty in essence.RStore consumes four to eight complete PSEG, depends on the data protection grade.Do not have the bar of redundance to cut RStore consumption 4 PSEG (4-2048 kilobyte PSEG=8 megabyte), the RStore with 4+1 parity checking consumes 5 PSEG, and eight PSEG of mirror image RStore consumption, realizes the virtual address space of 8 megabyte.

RStore and RAID dish collection are similar, and difference is that it comprises PSEG rather than physical disks.RStore is less than traditional RAID storage volume, and therefore, opposite with single RAID storage volume in the legacy system, given LUN 102 will comprise a plurality of RStore.

Consider that driver 405 can add or deletion from LDAD 103 in time.Add driver and mean that available data can more distribute on the multiple driver, the deletion driver means that then available data must move and fill the capacity that remains on the driver from the driver that withdraws from.This data migtation generally is called " evenly adjusting ".Evenly adjust the data of attempting the given LUN 102 of distribution on phisical drive as much as possible.Evenly the basic purpose of adjusting is the physical allocation of the represented memory block of each LUN 102 of distributing, make on the given physical disks for the use amount of given Logical Disk and that physical volume portion in can be used for being assigned to the physical storage area total amount of given Logical Disk proportional.

By data are copied to another from a PSEG, and then change data among the suitable RSD to show new membership, existing RStore can be revised as and use new PSEG.The follow-up RStore that creates in RSS will use new member automatically.Similarly, can be by data be copied to sky PSEG from filling PSEG, and change the data among the LMAP 502 so that the new PSEG composition of reflection RSD is eliminated PSEG.Like this, the relation between the logical expressions of physical storage area and memory block can be managed and be upgraded continuously, thereby reflects current storage environment in the sightless mode of user.

The snapshot difference file

In one aspect, this system configuration becomes to be implemented in the file that is called snapshot difference file or snapshot difference object herein.The snapshot difference file is to be designed to entity that some characteristic of snapshot (promptly when not changing data in the life cycle of snapshot difference with the volumetric efficiency of succession and precursor person's file-sharing data) is combined with the time response of journal file.The snapshot difference file also can be used in combination with basic snapshot clone and other snapshot difference, checks that data pass through the ability of the different copies of time so that provide.The snapshot difference file is also caught all new datas at the LUN that begins with certain time point, up to decision deactivation snapshot difference, and begins new snapshot difference.

The snapshot difference file can similarly constitute with snapshot.The snapshot difference can adopt to snapshot in the similar metadata structure of employed metadata structure, make the snapshot document can be in due course and precursor person LUN shared data, but when data appear at the cycle of activity of snapshot difference time of arrival, comprise unique or different data.Succession's snapshot difference can be quoted data among precursor person's snapshot difference or the precursor person LUN via identical mechanism.

As an example, suppose that LUN A was movable before at 1 in afternoon on September 12nd, 2004.The snapshot difference 1 of LUN A is movable to 2 pm after at 1 in the afternoon on September 12nd, 2004.The snapshot difference 2 of LUN A from after 2 pm on September 12nd, 2004 at 3 in afternoon be movable.The snapshot difference 1 of LUN A can adopt identical virtual index of metadata method to come access with data in each of snapshot difference 2.Snapshot difference 1 comprises the unique data that changes (with the granularity of employed index scheme) after at 1 in afternoon to 2 pm, and shares all other data with LUN A.Snapshot difference 2 comprises after 2 pm to 3 unique data that changed in afternoon, and shares all other data with snapshot difference 1 or LUN A.This The data is called the above-mentioned index of snapshot tree, a shared position scheme is come access.Therefore, be held over time-the LUN A view of data before at 1 in afternoon, 2 pm and before the snapshot difference 1 and the LUNA view of data, at 3 in afternoon and before snapshot difference 2 and snapshot difference 1 and LUN A view.Perhaps, split time view, snapshot difference 2 views of 3 data from 1 in afternoon to snapshot difference 1 view of the data of 2 pm or from 2 pm to afternoon.

Therefore, the similarity of snapshot difference and journal file is, the snapshot difference file is with data and time correlation (being the new datas of they collections from time a to time b), structurally similar snapshot of while (being the characteristic that they have snapshot, the ability that the speed of data access and space efficiency and maintenance just changes in time).

By crucial snapshot characteristic and structure are combined with the journal file time model, the snapshot difference can be used to provide high incremental backup of time maintenance, the simple space efficiency of synchronous mirror function, data all the time and powerful instantaneous recovery mechanism.

Fig. 7 is the signal high level specification that combines the storage data volume architecture of snapshot difference file.With reference to Fig. 7, it can be prenormalization snapshot clone or postnormalization snapshot clone's snapshot clone 720 that source book 710 is copied to.

Term as used herein " prenormalization snapshot clone " referred to before snapshot clone is cut apart from source book 710 the snapshot clone synchronous with source book 710.Prenormalization snapshot clone is illustrated in the time point copy of snapshot clone from the source book in the moment that source book separates.On the contrary, the postnormalization snapshot is cloned on the particular point in time to be created, but the complete separate copy of the data in the source book 710 is just finished after time point after a while.

The snapshot difference file is created on particular point in time and is activated, and subsequently, all I/O operations that influence the data in the source book 710 copy to movable snapshot difference file simultaneously.Put or when reaching certain threshold (for example when the snapshot difference file reaches pre-sizing), can close the snapshot difference file, and can activate another snapshot difference file in expeced time.After snapshot difference file 730,732,734 was deactivated, it can merge among the snapshot clone 720.In addition, the snapshot difference file can backup to tape drive, as tape drive 742,744,746.

In one implementation, snapshot difference file and snapshot clone, be created simultaneously and activate as snapshot clone 720 establishment.At the I/O of source book 710 operation copy to movable snapshot difference file, as snapshot difference file 730.

Illustrate in greater detail the snapshot difference file with reference to Fig. 8, Fig. 9 a-9b and Figure 10-13.Fig. 8 and Fig. 9 a-9b are the illustrative of the memory assignment map of snapshot difference file.Briefly with reference to Fig. 8, in one implementation, the memory mapped of snapshot difference file is in 800 beginnings of Logical Disk cell list, but it be the mapping number consecutively a plurality of Logical Disk status blocks (LDSB), be the array of the data structure of LDSB0, LDSB1...LDSB N.Each LDSB comprises the pointer that points to LMAP, the pointer that points to precursor person and succession LDSB.LMAP pointed LMAP mapping (enum) data structure, as mentioned above, it finally is mapped to PSEG (perhaps being mapped to the dish in the non-virtualization system).Precursor person is used for following the tracks of basic snapshot clone and relevant snapshot difference thereof with succession LDSB field.Basic snapshot clone is represented by the LDSB that does not have precursor person, and movable snapshot difference is represented by the LDSB that does not have the succession.

Fig. 9 a illustrates the memory mapped of snapshot difference file, wherein is provided with the shared position of RSD.Therefore, represent the LMAP 910 structure mapping RSD 915 of snapshot difference, RSD 915 is mapped to the LMAP 920 represented precursor person's snapshot difference or the basic snapshots of different pieces of information structure again and clones.This shows that LMAP 910 is successions of LMAP 920 and shares its data with LMAP 920.LMAP 920 is mapped to RSD 925, and RSD 925 is mapped to RSS930 again, and RSS 930 is mapped to physical disk space 935 (perhaps being mapped to the PSEG in the virtual storage system).Fig. 9 b illustrates the memory mapped of snapshot difference file, and the shared position of RSD wherein is not set, and promptly it is not shared.LMAP 950 is mapped to RSD 955, and RSD 955 is mapped to RSS 960 again, and RSS 960 is mapped to physical disk space 965 (perhaps being mapped to the PSEG in the virtual storage system).

Figure 10-the 13rd, process flow diagram illustrates the operation that is respectively applied in the demonstration methods that the snapshot difference is created, reads, write or merges.In the following description, everybody will appreciate that each frame of flowchart text and the combination of the frame in the flowchart text can realize by computer program instructions.These computer program instructions can be loaded into computing machine or other programmable device to produce a kind of machine, the parts of the function that the one or more frames that make the instruction that moves on processor or other programmable device create to be used for realization flow figure are specified.These computer program instructions also can be stored in the computer-readable memory, they can instruct computing machine or other programmable device with ad hoc fashion work, make the instruction of storing in the computer-readable memory produce a kind of manufacturing a product, the instruction unit of specified function in the one or more frames comprising realization flow figure.Computer program instructions also can be loaded into computing machine or other programmable device, the sequence of operations step is carried out on computing machine or other programmable device, thereby produce computer implemented process, make the instruction of on computing machine or other programmable device, carrying out be provided for the step of function specified in one or more frames of realization flow figure.

Therefore, the frame support of the flowchart text step combination that is used to carry out the unit construction of appointed function and is used to carry out appointed function.The combination that it can also be appreciated that each frame of flowchart text and the frame in the flowchart text can be by carrying out realizing based on the computer system of specialized hardware or the combination of specialized hardware and computer instruction of appointed function or step.

Figure 10 is a process flow diagram, and the operation of the demonstration methods that is used for creating the snapshot difference file is described.The operation of Figure 10 can respond and receive the request of creating the snapshot difference file and carry out in the suitable processor of array control unit in storage system for example etc.With reference to Figure 10, in operation 1010, create new LDSB, represent new snapshot difference.Referring again to Fig. 8 and suppose that LDSB0 to LDSB 3 has been assigned with, the new LDSB that is numbered LDSB 4 is created in operation 1010.At operation 1015-1020, LDSB succession's pointer is traveled through, and the LDSB that clones with snapshot begins, till running into sky succession pointer.When running into sky succession pointer, null pointer resets to and points to newly-built LDSB (operation 1025).Therefore, under situation shown in Figure 8, succession's pointer travels through to the LDSB3 with sky succession pointer from LDSB 0 to LDSB2 again.Operation 1025 resets to the succession's pointer among the LDSB 3 and points to LDSB4.Then, control forwards operation 1030 to, and wherein, precursor person's pointer of new LDSB is set up.Under situation shown in Figure 8, precursor person's pointer of LDSB 4 is set to point to LDSB 3.The high-level data mapping graph of the operative configuration snapshot difference file of Figure 10.Low level data mapping (that is, from LMAP to PSEG or physical disks section) can be carried out according to the above description that provides.

Figure 11 is a process flow diagram, and the operation that is used for carrying out the demonstration methods of read operation under the environment that utilizes one or more snapshot difference file is described.With reference to Figure 11, in operation 1110, for example the array control unit in storage system receives read request.In a demonstration realized, read request can be produced by principal computer, and can identify LBA (Logical Block Addressing) (LBA) in the storage system that will be read or another mark of address.In operation 1115, determine that whether read request is at the snapshot difference file.In a demonstration realized, the snapshot difference file can be assigned with specific LBA and/or the LD identifier that can be used to operate desired judgement in 1115.

If determine that in operation 1115 read request is not at the snapshot difference file, then control forwards operation 1135 to, and read request can be carried out according to the LD that course of normal operation is identified from read request.On the contrary, if determine that in operation 1115 read request is at the snapshot difference file, then executable operations 1120-1130 has the LBA that the snapshot difference file is identified to search in the read request now thereby travel through.

In operation 1120, movable snapshot difference file is examined, so that determine whether related with the LBA that identified in the read request shared be set up.Be not provided with if share the position, this shows that movable snapshot difference file comprises new data in identify LBA, and then control forwards to and operates 1135, and the LBA execution in the read request snapshot difference file that can be identified from read request.

On the contrary, if in operation 1120 shared position is not set, then control forwards operation 1125 to, determines therein whether the precursor person of movable snapshot difference file is another snapshot difference file.In a demonstration realized, this LDSB that can be identified by precursor person's pointer of analytic activity snapshot difference determined, as shown in Figure 8.If precursor person is not the snapshot difference file, then control forwards operation 1135 to, and read request can be carried out according to the LD that course of normal operation identifies from read request.On the contrary, if determine that in operation 1125 read request is at the snapshot difference file, executable operations 1125-1130 then, thereby the existing snapshot difference file of traversal, up at snapshot difference file or the location identifies in the read request in LD LBA, and LBA is read (operation 1135) and returns to requesting host (operation 1140).

Figure 12 is a process flow diagram, and the operation that is used for carrying out the demonstration methods of write operation under the environment that utilizes one or more snapshot difference file is described.With reference to Figure 12, in operation 1210, for example the array control unit in storage system receives write request.In a demonstration realized, write request can be produced by principal computer, and can identify write operation at storage system in LBA (Logical Block Addressing) (LBA) or another mark of address.In operation 1215, determine that whether write request is at the snapshot difference file.In a demonstration realized, the snapshot difference file can be assigned with specific LBA and/or the LD identifier that can be used to operate desired judgement in 1215.

If determine that in operation 1215 write request is not at the snapshot difference file, then control forwards operation 1245 to, and write request is carried out at the LD that identifies in the write request according to course of normal operation, and confirms to be returned to principal computer (operation 1255).On the contrary, if determine write request at the snapshot difference file in operation 1215, then executable operations 1220-1230 has the LBA that the snapshot difference file is identified to search in the write request now thereby travel through.

In operation 1220, movable snapshot difference file is examined, so that determine whether related with the LBA that identified in the read request shared be set up.Be not provided with if share the position, this shows that movable snapshot difference file comprises new data in sign LBA, and then control forwards operation 1250 to, and write request can be carried out for the LBA in the snapshot difference file that identifies in the write request.Be appreciated that write operation can only rewrite the LBA that is changed by write operation or comprise the whole RSEG of the LBA that write operation changes, depend on the configuration of system.

On the contrary, if in operation 1220 shared position is not set, then control forwards operation 1225 to, determines therein whether the precursor person of movable snapshot difference file is another snapshot difference file.In a demonstration realized, this LDSB that can be identified by precursor person's pointer of analytic activity snapshot difference determined, as shown in Figure 8.If precursor person is not the snapshot difference file, then control forwards operation 1235 to, and the LD that the RSEG related with the LBA that identifies in the write request can identify from write request copies in the impact damper.Then, control forwards operation 1240 to, and the I/O data in the write request merge in the impact damper.Then, control forwards operation 1250 to, and the I/O data are written into movable snapshot difference file, and in operation 1255, confirms to be returned to main frame.

On the contrary, if determine write request at the snapshot difference file in operation 1225, then executable operations 1225-1230 has the snapshot difference file now thereby travel through, up to locate the LBA that identifies in the write request in the snapshot difference file or in LD.Then, executable operations 1235-1250, thus the RSEG that write operation is changed copies to movable snapshot difference file.

As mentioned above, in one implementation, the snapshot difference file can be a time restriction, that is, the snapshot difference file can be activated at particular point in time, and can be deactivated at particular point in time.Figure 13 is a process flow diagram, and the operation that is used for the snapshot difference file is merged to the associated snapshot clone's of Logical Disk, for example snapshot difference demonstration methods is described.The operation of Figure 13 can be used as the background process periodic operation, perhaps can be triggered by particular event or sequence of events.

At this moment process receives the request that merges the snapshot difference file in operation 1310 beginnings.In a demonstration realized, merging request can be produced by principal computer, and can identify one or more snapshot difference file and snapshot difference file and will merge to wherein snapshot clone.

In operation 1315, search " the earliest " snapshot difference file.In a demonstration realizes, precursor person/succession's pointer track that can be by following the LDSB mapping graph, have till the LDSB of the precursor person's pointer that is mapped to the snapshot clone up to the location, locate snapshot difference the earliest.Referring again to Fig. 8 and suppose that LDSB 4 is movable snapshot difference file, the precursor person of LDSB 4 is LDSB 3.The precursor person of LDSB 3 is LDSB 2, and the precursor person of LDSB 2 is the LDSB 0 as the snapshot clone.Therefore, LDSB 2 expression " the earliest " snapshot difference file, it will merge among the snapshot clone.

The iterative loop that operation 1320 is initiated by each RSEG among each RSTORE that shines upon in the snapshot difference file.If in operation 1325, no longer include the RSEG that will analyze among the RSTORE, then control forwards operation 1360 to, and it determines whether the RSTORE that exists other to analyze.

If operating 1325, other RSEG that in RSTORE, will analyze in addition, then control forwards operation 1330 to, determines whether that therein RSEG is provided with the succession shares position or the shared position of precursor person.Any is set up in the position if these are shared, and then needs to merge the data among the RSEG, and therefore control forwards operation 1355 to.

On the contrary, do not share the position if be provided with in operation 1330, then control forwards operation 1335 to, and reads RSEG, and the data among the RSEG be replicated (operation 1340) to precursor person, be the respective memory unit among the snapshot clone.In operation 1345, share the position and in the RSEG of the snapshot difference that merges, reset.If in operation 1355, the RSEG that will analyze in addition among the RSTORE, then control forwards operation 1330 again to.Operation 1330-1355 repeats, up to by analysis all RSEG among the RSTORE, control at this moment forwards operation 1360 to, it determines whether the RSTORE that will analyze in addition.If in operation 1360, the RSTORE that will analyze then controls and forwards operation 1325 again in addition, and it is to the circulation of selected RSTORE resume operations 1330 to 1355.

Operation 1325 to 1360 repeats, up in operation 1360, no longer including the RSTORE that will analyze, in this case, control forwards operation 1365 to, and the succession's pointer among the precursor person LDSB LDSB of snapshot cloning cognate (promptly with) is set to point to the succession of merged LDSB.In operation 1370, merged LDSB is set to " sky ", thereby stops merging the existence of LDSB effectively.This process can repeat, so that " the earliest " snapshot difference file is one after the other merged among the snapshot clone.This also discharges the snapshot difference LDSB that has merged and supplies to use again.

As herein described is the demonstration methods that is called the file structure of snapshot difference file and is used to create and use the snapshot difference file.In a demonstration realized, the snapshot difference file can be cloned in conjunction with the snapshot in the remote copy operation and be realized.Difference file can be created and activate simultaneously with snapshot clone's generation.The I/O operation of the data in the source book of change and snapshot cloning cognate is recorded in the movable snapshot difference file.Movable snapshot difference file can be closed at particular point in time or when satisfying the certain threshold related with the snapshot difference file.Another snapshot difference file can be activated simultaneously with closing existing snapshot difference file, and the snapshot difference file can adopt the pointer that shows the time relationship between the snapshot difference file to link.After the snapshot difference file had been closed, file can merge among the snapshot clone related with it.

Backup operation

In demonstration realized, the snapshot difference file was used in and realizes the incremental backup process in storage networking and/or the memory storage, and they are not only saved the space but also save time, because backup operation only need be made the copy to the change of set of source data.With reference to Figure 14 a kind of such realization is described, wherein, Figure 14 is a process flow diagram, and the operation in the demonstration methods that utilizes the snapshot difference file in backup operation is described.

The operation of Figure 14 can realize by computer program instructions.These computer program instructions can be loaded into computing machine or other programmable device to produce a kind of machine, the parts of the function that the one or more frames that make the instruction of carrying out on processor or other programmable device create to be used for realization flow figure are specified.These computer program instructions also can be stored in the computer-readable memory, they can instruct computing machine or other programmable device with ad hoc fashion work, make the instruction of storing in the computer-readable memory produce a kind of manufacturing a product, the instruction unit of specified function in the one or more frames comprising realization flow figure.Computer program instructions also can be loaded into computing machine or other programmable device, the sequence of operations step is carried out on computing machine or other programmable device, thereby produce computer implemented process, make the instruction of on computing machine or other programmable device, carrying out be provided for the step of function specified in one or more frames of realization flow figure.

Therefore, the frame support of the flowchart text step combination that is used to carry out the unit construction of appointed function and is used to carry out appointed function.Will appreciate that also each frame of flowchart text and the combination of the frame in the flowchart text can be by carrying out realizing based on the computer system of specialized hardware or the combination of specialized hardware and computer instruction of appointed function or step.

With reference to Figure 14,, produce the snapshot clone of source file in operation 1410.In operation 1415, the snapshot difference file is activated, and in operation 1420, the I/O operation that changes the data in the source book is recorded in the snapshot difference file.These operations can be carried out according to the above description that provides.

In operation 1425, produce snapshot clone's backup copies.This operation can respond by the user and carry out in the backup request of user interface input, and perhaps incident of automated back-up operation of for example being driven by timer of response and so on or response source book or snapshot clone reach specific size and carry out.Backup copies can be recorded on another disk drive, tape drive or other medium.Replicate run can realize by background process, makes that replicate run is that the user of storage system is sightless.

In operation 1435, movable snapshot difference file can be closed, and at this moment, new snapshot difference file is activated (operation 1440), and buttoned-up snapshot difference file can merge in the snapshot clone file, as mentioned above.

In operation 1443, produce the copy of snapshot difference file.This operation can respond by the user and carry out in the backup request of user interface input, and perhaps incident of automated back-up operation of for example being driven by timer of response and so on or response snapshot difference reach specific size and carry out.Before backup operation, the snapshot difference needs deactivated or closes, and another snapshot difference is activated.Backup copies can be recorded on another disk drive, tape drive or other medium.Replicate run can realize by background process, makes that replicate run is that the user of storage system is sightless.The so-called incremental backup of such backup, and in effective life cycle of snapshot difference file, carry out once usually.Adopt the unique aspect of such incremental backup of snapshot difference to be, it identifies under the reformed situation of what content need not applications or file system, and the ability that only backs up the content that has changed on the grain size category of employed virtual mapping is provided.

Operation 1430 to 1445 can be carried out ad infinitum repeatedly, continues to write down the I/O operation and preserve the copy of snapshot difference file in the snapshot difference file in suitable storage medium.Therefore, operation 1410 and 1415 can be carried out to produce snapshot clone and snapshot difference file at very first time point respectively.Operation 1325 can be carried out the backup copies of cloning with the generation snapshot at second time point, and operates 1443 and can carry out to produce the copy of snapshot difference file at the 3rd time point.The follow-up copy of snapshot difference can produce at follow-up time point.

Figure 14 illustrates the complete sequence that adopts the snapshot difference file to carry out the operation of incremental backup operations.Person of skill in the art will appreciate that, operate 1410 to 1420 and can independently realize, promptly as mentioned above.The operation of Figure 14 is suitable for coiling tape backup system most.

The backup operation that adopts the snapshot difference file is to save the space, because have only the change ability record in backup operation to source book.In addition, the snapshot difference file can be used for the automatic supervisory routine of backup operation.Figure 15 is a process flow diagram, and an operation that demonstration realizes of the method that is used for managing automatically backup operation is described.Such backup model is suitable for coiling the dish standby system most.

In operation 1510, receive the backup setting indicator signal.In one implementation, the backup setting indicator signal can be produced at the appropriate users interface by the user, and shows the threshold number of the snapshot difference file that will preserve.Thresholding can be expressed as file for example quantity, can distribute to the snapshot difference file storage space maximum or be expressed as time parameter.In operation 1515, this threshold number is determined from signal.

If in operation 1520, the current quantity of snapshot difference file is greater than quantity indicated in the backup setting indicator signal, and then control forwards operation 1525 to, and adopts said process that " the earliest " snapshot difference file is merged in the snapshot clone file.Operation 1520 to 1525 can be carried out repeatedly, up to the current quantity of snapshot difference file less than the backup setting indicator signal in indicated thresholding, at this moment, control forwards operation 1530 to, and produces the snapshot clone.

In operation 1533, current snapshot difference file is deactivated, and in operation 1535, new snapshot difference can be activated, and the I/O operation of source book is written into movable snapshot difference file (operation 1540).

The operation of Figure 15 allows the user of system to specify in the maximum quantity of the snapshot difference file that will preserve in the background copy operation.As an example, the storage system of using in user's possible configuration office, be set to open every day new snapshot difference file, and every day, backup copies was the snapshot difference file.The user also can specify the maximum quantity of seven snapshot difference file, makes in every day, and system produces the circulation copy of snapshot clone file every day.Snapshot difference the earliest can be re-circulated among the snapshot clone every day.Those skilled in the art can know that other configuration is available.

Though described scheme and process are described by architectural feature and/or the specific language of method operation, are appreciated that the theme that defines in the claims not necessarily is confined to described special characteristic or operation.On the contrary, special characteristic and operation require the preferred form of this theme of its rights and interests to come openly as realization.

Claims (8)

1. method of carrying out backup operation in storage networking comprises:
Produce the snapshot clone (720) of source tray volume (710) at very first time point;
Activate the first snapshot difference file (730) that logically is linked to described snapshot clone (720) simultaneously;
The I/O operation note that will change the data set in the described source tray volume (710) is in the described first snapshot difference file (730);
Close the described first snapshot difference file (730);
After described very first time point, produce described snapshot clone's (720) backup copies at second time point; And
After described second time point, produce the backup copies of the described first snapshot difference file (730) at the 3rd time point.
2. the method for claim 1 is characterized in that, also comprises:
Activate the second snapshot difference file (732) afterwards closing the described first snapshot difference file (730); And
The I/O operation note that will change the data set in the source tray volume (710) is in the described second snapshot difference file (732).
3. method as claimed in claim 2 is characterized in that, also is included in the backup copies that described the 3rd time point the 4th time point afterwards produces the described second snapshot difference file (732).
4. the method for claim 1 is characterized in that, the described first snapshot difference file (730) comprises and is used for the I/O operation that record carries out for described source tray volume (710) and is used to write down data field with time of each I/O operative association.
5. the method for claim 1 is characterized in that, after described very first time point, the backup copies that produces described snapshot clone (720) at second time point comprises backup copies is write permanent storage media.
6. the method for claim 1 is characterized in that, after described second time point, the backup copies that produces the described first snapshot difference file (730) at the 3rd time point comprises backup copies is write permanent storage media.
7. method as claimed in claim 2 is characterized in that, also is included in to close the described first snapshot difference file (730) that the described first snapshot difference file (730) will close afterwards and merge among the described snapshot clone (720).
8. method as claimed in claim 7 is characterized in that, also is included in to carry out the backup copies that described union operation produces described snapshot clone (720) afterwards.
CNB2005101188875A 2004-11-02 2005-11-01 Incremental backup operations in storage networks CN100419664C (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US10/979,395 US20060106893A1 (en) 2004-11-02 2004-11-02 Incremental backup operations in storage networks
US10/979395 2004-11-02

Publications (2)

Publication Number Publication Date
CN1770088A CN1770088A (en) 2006-05-10
CN100419664C true CN100419664C (en) 2008-09-17

Family

ID=35825391

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2005101188875A CN100419664C (en) 2004-11-02 2005-11-01 Incremental backup operations in storage networks

Country Status (4)

Country Link
US (1) US20060106893A1 (en)
EP (1) EP1653358A3 (en)
CN (1) CN100419664C (en)
SG (1) SG121964A1 (en)

Families Citing this family (56)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060253497A1 (en) * 2005-05-03 2006-11-09 Bulent Abali System and method for associating computational procedures with stored data objects
US20070226396A1 (en) * 2006-03-09 2007-09-27 International Business Machines Corporation System and method for backing up and recovering data in a computer system
US8025650B2 (en) * 2006-06-12 2011-09-27 Wound Care Technologies, Inc. Negative pressure wound treatment device, and methods
US7853567B2 (en) * 2006-08-04 2010-12-14 Apple Inc. Conflict resolution in recovery of electronic data
US7809688B2 (en) * 2006-08-04 2010-10-05 Apple Inc. Managing backup of content
US8311988B2 (en) 2006-08-04 2012-11-13 Apple Inc. Consistent back up of electronic information
US7856424B2 (en) * 2006-08-04 2010-12-21 Apple Inc. User interface for backup management
US7853566B2 (en) 2006-08-04 2010-12-14 Apple Inc. Navigation of electronic backups
US20080034017A1 (en) * 2006-08-04 2008-02-07 Dominic Giampaolo Links to a common item in a data structure
US7860839B2 (en) * 2006-08-04 2010-12-28 Apple Inc. Application-based backup-restore of electronic information
US20080034004A1 (en) * 2006-08-04 2008-02-07 Pavel Cisler System for electronic backup
US8166415B2 (en) 2006-08-04 2012-04-24 Apple Inc. User interface for backup management
US9009115B2 (en) * 2006-08-04 2015-04-14 Apple Inc. Restoring electronic information
US7809687B2 (en) * 2006-08-04 2010-10-05 Apple Inc. Searching a backup archive
US20080126442A1 (en) * 2006-08-04 2008-05-29 Pavel Cisler Architecture for back up and/or recovery of electronic data
US8370853B2 (en) * 2006-08-04 2013-02-05 Apple Inc. Event notification management
US20080091744A1 (en) * 2006-10-11 2008-04-17 Hidehisa Shitomi Method and apparatus for indexing and searching data in a storage system
US8589341B2 (en) * 2006-12-04 2013-11-19 Sandisk Il Ltd. Incremental transparent file updating
US7865473B2 (en) * 2007-04-02 2011-01-04 International Business Machines Corporation Generating and indicating incremental backup copies from virtual copies of a data set
US8725965B2 (en) 2007-06-08 2014-05-13 Apple Inc. System setup for electronic backup
US8010900B2 (en) * 2007-06-08 2011-08-30 Apple Inc. User interface for electronic backup
AT519159T (en) * 2007-06-08 2011-08-15 Apple Inc Applications-based backup restore of electronic information
US8468136B2 (en) * 2007-06-08 2013-06-18 Apple Inc. Efficient data backup
US8745523B2 (en) * 2007-06-08 2014-06-03 Apple Inc. Deletion in electronic backups
US8307004B2 (en) 2007-06-08 2012-11-06 Apple Inc. Manipulating electronic backups
US8429425B2 (en) * 2007-06-08 2013-04-23 Apple Inc. Electronic backup and restoration of encrypted data
US20080307017A1 (en) 2007-06-08 2008-12-11 Apple Inc. Searching and Restoring of Backups
US8099392B2 (en) 2007-06-08 2012-01-17 Apple Inc. Electronic backup of applications
CN100478904C (en) * 2007-07-18 2009-04-15 华为技术有限公司 Method and device for protecting snapshot
US20090210462A1 (en) * 2008-02-15 2009-08-20 Hitachi, Ltd. Methods and apparatus to control transition of backup data
US8135676B1 (en) * 2008-04-28 2012-03-13 Netapp, Inc. Method and system for managing data in storage systems
US20110093437A1 (en) * 2009-10-15 2011-04-21 Kishore Kaniyar Sampathkumar Method and system for generating a space-efficient snapshot or snapclone of logical disks
US8190574B2 (en) * 2010-03-02 2012-05-29 Storagecraft Technology Corporation Systems, methods, and computer-readable media for backup and restoration of computer information
US8548944B2 (en) 2010-07-15 2013-10-01 Delphix Corp. De-duplication based backup of file systems
CN102375696A (en) * 2010-08-20 2012-03-14 英业达股份有限公司 Data storage system and data access method by utilizing virtual disk
US9015119B2 (en) 2010-10-26 2015-04-21 International Business Machines Corporation Performing a background copy process during a backup operation
US8468174B1 (en) 2010-11-30 2013-06-18 Jedidiah Yueh Interfacing with a virtual database system
TW201227268A (en) * 2010-12-20 2012-07-01 Chunghwa Telecom Co Ltd Data backup system and data backup and retrival method
US8943026B2 (en) 2011-01-14 2015-01-27 Apple Inc. Visual representation of a local backup
US8984029B2 (en) 2011-01-14 2015-03-17 Apple Inc. File system management
US8639976B2 (en) * 2011-02-15 2014-01-28 Coraid, Inc. Power failure management in components of storage area network
CN102147754A (en) * 2011-04-01 2011-08-10 奇智软件(北京)有限公司 Automatic backup method and device for driver
WO2014115314A1 (en) * 2013-01-25 2014-07-31 株式会社 東芝 Backup storage system that backs up data and backup storage device and method
CN103164294A (en) * 2013-01-30 2013-06-19 浪潮(北京)电子信息产业有限公司 System, device and method achieving restoring points of computer
US10635638B2 (en) * 2013-03-13 2020-04-28 Ivanti Us Llc Systems, methods and media for deferred synchronization of files in cloud storage client device
CN103500130B (en) * 2013-09-11 2015-07-29 上海爱数软件有限公司 A kind of method that two-node cluster hot backup data are backed up in realtime
CN103581339A (en) * 2013-11-25 2014-02-12 广东电网公司汕头供电局 Storage resource allocation monitoring and processing method based on cloud computing
CN103645970B (en) * 2013-12-13 2017-04-19 华为技术有限公司 Realizing method and device for de-weighting increments among multiple snapshots for remote copy
US9645739B2 (en) * 2014-09-26 2017-05-09 Intel Corporation Host-managed non-volatile memory
CN104375904A (en) * 2014-10-30 2015-02-25 浪潮电子信息产业股份有限公司 Disaster recovery backup method based on snapshot differentiation data transmission
US9870289B2 (en) * 2014-12-12 2018-01-16 Ca, Inc. Notifying a backup application of a backup key change
US9880904B2 (en) * 2014-12-12 2018-01-30 Ca, Inc. Supporting multiple backup applications using a single change tracker
CN106325769B (en) * 2016-08-19 2019-05-28 华为技术有限公司 A kind of method and device of data storage
CN107291883B (en) * 2017-06-20 2020-09-18 苏州浪潮智能科技有限公司 Snapshot method and device of distributed storage system
CN107526656B (en) * 2017-08-31 2020-11-20 苏州浪潮智能科技有限公司 Cloud restoration method and device
CN111522501A (en) * 2019-02-01 2020-08-11 杭州海康威视数字技术股份有限公司 Disk array space dividing method and device, electronic equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6003044A (en) * 1997-10-31 1999-12-14 Oracle Corporation Method and apparatus for efficiently backing up files using multiple computer systems
US20030079102A1 (en) * 2001-06-01 2003-04-24 Lubbers Clark E. System and method for generating point in time storage copy
US6560615B1 (en) * 1999-12-17 2003-05-06 Novell, Inc. Method and apparatus for implementing a highly efficient, robust modified files list (MFL) for a storage system volume
US6594744B1 (en) * 2000-12-11 2003-07-15 Lsi Logic Corporation Managing a snapshot volume or one or more checkpoint volumes with multiple point-in-time images in a single repository
CN1452737A (en) * 2000-04-24 2003-10-29 微软公司 Method and apparatus for providing volume snapshot dependencies in computer system
CN1530841A (en) * 2003-03-13 2004-09-22 威达电股份有限公司 Memory system and method by snapshot back-up function

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4751702A (en) * 1986-02-10 1988-06-14 International Business Machines Corporation Improving availability of a restartable staged storage data base system that uses logging facilities
US5857208A (en) * 1996-05-31 1999-01-05 Emc Corporation Method and apparatus for performing point in time backup operation in a computer system
US7203732B2 (en) * 1999-11-11 2007-04-10 Miralink Corporation Flexible remote data mirroring
US6665815B1 (en) * 2000-06-22 2003-12-16 Hewlett-Packard Development Company, L.P. Physical incremental backup using snapshots
US20020104008A1 (en) * 2000-11-30 2002-08-01 Cochran Robert A. Method and system for securing control-device-lun-mediated access to luns provided by a mass storage device
US6629203B1 (en) * 2001-01-05 2003-09-30 Lsi Logic Corporation Alternating shadow directories in pairs of storage spaces for data storage
US6594745B2 (en) * 2001-01-31 2003-07-15 Hewlett-Packard Development Company, L.P. Mirroring agent accessible to remote host computers, and accessing remote data-storage devices, via a communcations medium
US6606690B2 (en) * 2001-02-20 2003-08-12 Hewlett-Packard Development Company, L.P. System and method for accessing a storage area network as network attached storage
US6697881B2 (en) * 2001-05-29 2004-02-24 Hewlett-Packard Development Company, L.P. Method and system for efficient format, read, write, and initial copy processing involving sparse logical units
US6728848B2 (en) * 2001-06-11 2004-04-27 Hitachi, Ltd. Method and system for backing up storage system data
US6895467B2 (en) * 2001-10-22 2005-05-17 Hewlett-Packard Development Company, L.P. System and method for atomizing storage
US7296125B2 (en) * 2001-11-29 2007-11-13 Emc Corporation Preserving a snapshot of selected data of a mass storage system
US20030120676A1 (en) * 2001-12-21 2003-06-26 Sanrise Group, Inc. Methods and apparatus for pass-through data block movement with virtual storage appliances
US7036043B2 (en) * 2001-12-28 2006-04-25 Storage Technology Corporation Data management with virtual recovery mapping and backward moves
US6763436B2 (en) * 2002-01-29 2004-07-13 Lucent Technologies Inc. Redundant data storage and data recovery system
US6829617B2 (en) * 2002-02-15 2004-12-07 International Business Machines Corporation Providing a snapshot of a subset of a file system
US7143307B1 (en) * 2002-03-15 2006-11-28 Network Appliance, Inc. Remote disaster recovery and data migration using virtual appliance migration
US7844577B2 (en) * 2002-07-15 2010-11-30 Symantec Corporation System and method for maintaining a backup storage system for a computer system
US7100089B1 (en) * 2002-09-06 2006-08-29 3Pardata, Inc. Determining differences between snapshots
JP4402992B2 (en) * 2004-03-18 2010-01-20 株式会社日立製作所 Backup system, method and program

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6003044A (en) * 1997-10-31 1999-12-14 Oracle Corporation Method and apparatus for efficiently backing up files using multiple computer systems
US6560615B1 (en) * 1999-12-17 2003-05-06 Novell, Inc. Method and apparatus for implementing a highly efficient, robust modified files list (MFL) for a storage system volume
CN1452737A (en) * 2000-04-24 2003-10-29 微软公司 Method and apparatus for providing volume snapshot dependencies in computer system
US6594744B1 (en) * 2000-12-11 2003-07-15 Lsi Logic Corporation Managing a snapshot volume or one or more checkpoint volumes with multiple point-in-time images in a single repository
US20030079102A1 (en) * 2001-06-01 2003-04-24 Lubbers Clark E. System and method for generating point in time storage copy
CN1530841A (en) * 2003-03-13 2004-09-22 威达电股份有限公司 Memory system and method by snapshot back-up function

Also Published As

Publication number Publication date
SG121964A1 (en) 2006-05-26
EP1653358A2 (en) 2006-05-03
CN1770088A (en) 2006-05-10
US20060106893A1 (en) 2006-05-18
EP1653358A3 (en) 2009-10-28

Similar Documents

Publication Publication Date Title
US10503427B2 (en) Synchronously replicating datasets and other managed objects to cloud-based storage systems
US9930118B2 (en) Systems and methods for granular resource management in a storage network
US9606740B2 (en) System, method and computer program product for synchronizing data written to tape including writing an index into a data partition
CN105190622B (en) Quick collapse for distributed data base system restores
CN105122241B (en) With database engine and the Database Systems for being independently distributed formula storage service
CN105210062B (en) System scope checkpoint for distributed data base system avoids
US9740560B2 (en) Failure resilient distributed replicated data storage system
JP6050316B2 (en) Method and network storage server used in data storage system
CN102460439B (en) Data distribution through capacity leveling in a striped file system
Wilkes et al. The HP AutoRAID hierarchical storage system
US8843447B2 (en) Resilient distributed replicated data storage system
KR100376747B1 (en) Method, system and program for maintaninig data consistency across groups of storage areas
US8161321B2 (en) Virtual machine-based on-demand parallel disaster recovery system and the method thereof
CN101495970B (en) System and method for high performance enterprise data protection
US7949512B2 (en) Systems and methods for performing virtual storage operations
US7343460B2 (en) Data storage system and process
US8095852B2 (en) Data recorder
US7340640B1 (en) System and method for recoverable mirroring in a storage environment employing asymmetric distributed block virtualization
US8060713B1 (en) Consolidating snapshots in a continuous data protection system using journaling
US8762672B2 (en) Storage system and storage migration method
EP2558949B1 (en) Express-full backup of a cluster shared virtual machine
US7870353B2 (en) Copying storage units and related metadata to storage
CN102422277B (en) Converting luns into files or files into luns in real time
US7035880B1 (en) Modular backup and retrieval system used in conjunction with a storage area network
JP4791051B2 (en) Method, system, and computer program for system architecture for any number of backup components

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C41 Transfer of patent application or patent right or utility model
TR01 Transfer of patent right

Effective date of registration: 20170123

Address after: American Texas

Patentee after: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Address before: American Texas

Patentee before: Hewlett Packard Development Co.

CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20080917

Termination date: 20161101